WEBVTT 1 00:00:02.970 --> 00:00:06.570 Irene Palacios: I want to show you on Marchel 27, 2 00:00:08.780 --> 00:00:15.170 Irene Palacios: which is the Chi-square test for independence. This is pronounced Chi, the CHI. 3 00:00:15.690 --> 00:00:18.120 Irene Palacios: And I guess this is quiz number 3. 4 00:00:18.770 --> 00:00:22.010 Irene Palacios: There is a table. 5 00:00:23.200 --> 00:00:26.299 Irene Palacios: and it might be your question 3, or 6 00:00:26.420 --> 00:00:29.059 Irene Palacios: you might get another question for you. But 7 00:00:29.250 --> 00:00:31.220 Irene Palacios: this would be similar. 8 00:00:31.880 --> 00:00:35.400 Irene Palacios: Okay? So we have a table here. The data gives us 9 00:00:35.940 --> 00:00:36.660 Irene Palacios: go ahead. 10 00:00:37.480 --> 00:00:38.869 Irene Palacios: A table with 2 11 00:00:38.900 --> 00:00:42.600 Irene Palacios: categorical variables. One of the variables 12 00:00:42.620 --> 00:00:47.519 Irene Palacios: is whether or not a fireman participated in the 9 11 rescue. 13 00:00:47.650 --> 00:00:51.270 Irene Palacios: and then the other. Variable was their risk 14 00:00:52.160 --> 00:01:03.550 Irene Palacios: problems. So no risk moderate to severe risk for alcohol problems. So you can actually put this on step, Prince, the directions are here. 15 00:01:03.900 --> 00:01:09.829 Irene Palacios: And so if you click on the directions, this is just an example. The data, of course, is different. 16 00:01:10.470 --> 00:01:13.189 Irene Palacios: But you are able to change the 17 00:01:13.470 --> 00:01:19.759 Irene Palacios: title on staff front. So what I'm going to do is I'm just going to use these instructions right here. 18 00:01:20.880 --> 00:01:23.710 Irene Palacios: mainly down here, is what we see at all. 19 00:01:24.000 --> 00:01:29.689 Irene Palacios: and I'm going to enter this table into stack crunch again. The directions I did get them from here 20 00:01:29.980 --> 00:01:31.040 Irene Palacios: there's a link. 21 00:01:32.030 --> 00:01:35.429 Irene Palacios: So if you open your stack crunch. 22 00:01:36.260 --> 00:01:38.219 Irene Palacios: I went ahead and 23 00:01:45.270 --> 00:01:46.100 Irene Palacios: Buddy 24 00:01:52.400 --> 00:01:53.430 Irene Palacios: stay home. 25 00:01:54.290 --> 00:01:58.459 Irene Palacios: They try to work on this so I wouldn't make this video too long. 26 00:01:58.680 --> 00:02:01.320 Irene Palacios: But notice that for the first 27 00:02:01.340 --> 00:02:05.409 Irene Palacios: column I can just change these these variables here. 28 00:02:06.640 --> 00:02:08.369 Irene Palacios: you can just click on it. 29 00:02:08.620 --> 00:02:11.890 Irene Palacios: and you can change it 30 00:02:11.910 --> 00:02:19.539 Irene Palacios: right. And you can also make them a little bit wider. So you can actually fit anything you want here. 31 00:02:20.320 --> 00:02:24.060 Irene Palacios: So you can type anything you want here. 32 00:02:24.310 --> 00:02:26.110 Irene Palacios: Can you click on it again? 33 00:02:26.260 --> 00:02:27.820 Irene Palacios: We can type. 34 00:02:31.870 --> 00:02:37.299 Irene Palacios: and you can make that a little bit wider by just moving 35 00:02:37.740 --> 00:02:43.669 Irene Palacios: moving these right here, these dividers, and you've changed the name of your home. 36 00:02:43.890 --> 00:02:46.240 Irene Palacios: You can type anything you need to in here. 37 00:02:46.450 --> 00:02:49.810 Irene Palacios: So what I did is for this one. 38 00:02:51.280 --> 00:02:56.939 Irene Palacios: It looks like it is a whether or not they participated in the 9 11 rescue. 39 00:02:59.020 --> 00:03:01.359 Irene Palacios: And so that's what I did. 40 00:03:01.620 --> 00:03:04.889 Irene Palacios: Participant 9, 11. Rescue, yes or no. 41 00:03:06.320 --> 00:03:11.410 Irene Palacios: and they participate, or there, or they didn't, to put yes or no. And then for 42 00:03:11.550 --> 00:03:15.419 Irene Palacios: the other variable, it's the type of risk 43 00:03:15.500 --> 00:03:19.850 Irene Palacios: for alcohol problems they had so none or moderate to severe. 44 00:03:20.250 --> 00:03:23.569 Irene Palacios: So I just changed it up here. No risk. 45 00:03:23.800 --> 00:03:27.920 Irene Palacios: and then moderate to severe. And I just typed in the numbers 46 00:03:28.270 --> 00:03:30.099 Irene Palacios: 7, 93, 47 00:03:30.890 --> 00:03:32.350 Irene Palacios: 4, 41, 48 00:03:35.390 --> 00:03:45.670 Irene Palacios: 309, 1, 10. Notice that I just typed in the numbers here. I didn't do the row columns or the row totals. 49 00:03:45.840 --> 00:03:49.329 Irene Palacios: So I didn't do the row totals. I didn't do the row totals 50 00:03:49.400 --> 00:03:52.159 Irene Palacios: I just typed in the actual data. 51 00:03:52.400 --> 00:03:54.280 Irene Palacios: These 4 numbers right here. 52 00:03:55.800 --> 00:04:00.280 Irene Palacios: Stack French does all of the other stuff for you, and 53 00:04:01.950 --> 00:04:03.729 Irene Palacios: so if 54 00:04:04.330 --> 00:04:07.490 Irene Palacios: once you're done, once you've done with your table. 55 00:04:08.120 --> 00:04:11.260 Irene Palacios: you can just go to that. 56 00:04:13.200 --> 00:04:15.769 Irene Palacios: and then you're going to go to tables 57 00:04:16.850 --> 00:04:21.729 Irene Palacios: and then contingency. That's what we have here. And with summary 58 00:04:23.830 --> 00:04:30.510 Irene Palacios: alright. So then, what I did here is. I just followed the instructions that I have 59 00:04:30.810 --> 00:04:34.710 Irene Palacios: in my link within the actual problem. Remember. 60 00:04:34.730 --> 00:04:44.899 Irene Palacios: this link right here has directions. I just went ahead. I clicked it, and it says what to do. I I went to Stat tables contingency with Summary. 61 00:04:45.540 --> 00:04:47.460 Irene Palacios: So that's where I'm at right now. 62 00:04:48.780 --> 00:04:50.770 Irene Palacios: That gives me this table. 63 00:04:51.360 --> 00:04:55.040 Irene Palacios: Now it says, under the select columns. 64 00:04:56.270 --> 00:05:01.170 Irene Palacios: we want to click on each group associated with the response variable. 65 00:05:01.880 --> 00:05:04.719 Irene Palacios: So the response variable here 66 00:05:05.850 --> 00:05:07.190 Irene Palacios: and in this. 67 00:05:07.680 --> 00:05:09.690 Irene Palacios: whether they're drinking alcohol 68 00:05:11.090 --> 00:05:12.490 Irene Palacios: are the problems. 69 00:05:12.590 --> 00:05:14.400 Irene Palacios: their alcohol plans? 70 00:05:14.520 --> 00:05:16.420 Irene Palacios: That's their 71 00:05:17.040 --> 00:05:18.800 Irene Palacios: response mode. 72 00:05:19.410 --> 00:05:22.750 Irene Palacios: And then the explanatory variable. 73 00:05:23.570 --> 00:05:25.640 Irene Palacios: That's gonna be your role label. 74 00:05:26.220 --> 00:05:29.270 Irene Palacios: So the explanatory, variable. That's 75 00:05:29.360 --> 00:05:32.720 Irene Palacios: whether or not they participated in the 9 11 76 00:05:32.800 --> 00:05:33.890 Irene Palacios: rescue. 77 00:05:35.740 --> 00:05:39.589 Irene Palacios: And that's your row label. So does that explain 78 00:05:40.760 --> 00:05:45.579 Irene Palacios: what's happening with response variable whether they're having drinking. Certainly. 79 00:05:46.390 --> 00:05:55.169 Irene Palacios: Okay. So the row variant row labels is the explanatory variable, which is whether or not they participated in the menu skew 80 00:05:55.510 --> 00:05:59.939 Irene Palacios: and under columns. We're gonna click on 81 00:06:00.510 --> 00:06:01.650 Irene Palacios: down session 82 00:06:01.980 --> 00:06:05.159 Irene Palacios: usage. Okay? So let's go back 83 00:06:06.690 --> 00:06:16.720 Irene Palacios: alright. So here, you're gonna have to click and then control click, so that you can select both of them. So click and then control click. 84 00:06:17.240 --> 00:06:25.619 Irene Palacios: And then that's what selects both. So this is about the problems they're having with alcohol, either no risk or moderate severe. 85 00:06:26.100 --> 00:06:33.260 Irene Palacios: This is, that's your response. And this is your explanatory. That's whether or not they participated in 86 00:06:35.160 --> 00:06:38.419 Irene Palacios: alright. And I just go back here and continue 87 00:06:38.460 --> 00:06:39.640 Irene Palacios: reading. 88 00:06:39.700 --> 00:06:42.669 Irene Palacios: It says, under the display 89 00:06:44.820 --> 00:06:48.930 Irene Palacios: expected count, okay? So under display 90 00:06:50.520 --> 00:06:58.739 Irene Palacios: expected count. So that's what we did here. And then it's the Chi-square test for independence. And then I compute, and there I get my table 91 00:06:59.300 --> 00:07:00.950 Irene Palacios: had done it earlier. 92 00:07:01.340 --> 00:07:03.749 Irene Palacios: checking that I do have the same data. 93 00:07:04.190 --> 00:07:12.309 Irene Palacios: Okay? So if you were to calculate the expected values by hand, this is what you would get what you see here in parentheses. 94 00:07:12.390 --> 00:07:19.230 Irene Palacios: stack friendship already does that for us so for assuming that the 2 variables are independent. 95 00:07:19.440 --> 00:07:23.449 Irene Palacios: then what we see in parentheses is what we should have seen. 96 00:07:23.810 --> 00:07:36.190 Irene Palacios: so we should see 822 or 823 individuals who participated in the 9 11 rescue have no respir. 97 00:07:40.090 --> 00:07:41.420 Irene Palacios: while 98 00:07:43.050 --> 00:07:51.879 Irene Palacios: 279 of them who did participate in the 9 11 rescue would have moderate severe risk for alcohol. 99 00:07:52.300 --> 00:08:01.940 Irene Palacios: So what's in parentheses is what we would expect to see if the 2 variables were not related were independent of each other. 100 00:08:03.350 --> 00:08:28.910 Irene Palacios: Now you can notice a difference between what's in parentheses, which is our expected value, and what we actually saw. So 823 and 793, those are different. Right? 279 and 309. Those are different to the question is, are the differences significant? And are they statistically significant? 101 00:08:29.190 --> 00:08:36.459 Irene Palacios: Because if they are statistically significant, then we would reject the null, and we would reject independence. 102 00:08:36.770 --> 00:08:40.040 Irene Palacios: because the null is a statement of independence 103 00:08:40.070 --> 00:08:41.980 Irene Palacios: that the 2 are not related. 104 00:08:42.659 --> 00:08:47.810 Irene Palacios: Right? So, as I'm saying all that, I want you to remember that the null 105 00:08:47.830 --> 00:08:51.359 Irene Palacios: is the statement of independence. The 2 variables 106 00:08:51.370 --> 00:09:03.439 Irene Palacios: have no relationship. These are the notes from Module 27. So this is not the problem itself. It's just the setup. So the null is no relationship. The 2 variables are independent. 107 00:09:04.040 --> 00:09:10.619 Irene Palacios: See the null, the 2 variables are independent and the alternate. The 2 variables are dependent. 108 00:09:10.750 --> 00:09:12.519 Irene Palacios: So that's how we set it up 109 00:09:12.750 --> 00:09:17.939 Irene Palacios: again. This was just from my notes, as far as the setup is concerned. 110 00:09:18.680 --> 00:09:21.520 Irene Palacios: but it's not the exact problem 111 00:09:23.810 --> 00:09:24.620 Irene Palacios: alright. 112 00:09:25.460 --> 00:09:26.640 Irene Palacios: So 113 00:09:26.870 --> 00:09:32.609 Irene Palacios: what we find is that our Pw is really it's pretty small, less than 114 00:09:32.680 --> 00:09:35.549 Irene Palacios: point 0 5. What's last 1.0 1. 115 00:09:36.090 --> 00:09:38.710 Irene Palacios: So it is smaller than Alpha. 116 00:09:38.790 --> 00:09:42.599 Irene Palacios: So then we reject them all. So we're rejecting independence. 117 00:09:42.770 --> 00:09:44.849 Irene Palacios: and we're rejecting independence. 118 00:09:45.490 --> 00:09:51.360 Irene Palacios: We are saying that the 2 are related. They are dependent. So it's the alternate. 119 00:09:52.010 --> 00:09:53.440 Irene Palacios: That's what our data 120 00:09:55.550 --> 00:10:06.310 Irene Palacios: alright. So you can go back over here and enter all your information for the Chi squared p-value related to. I'm just going to submit the quiz. I did do the whole thing. 121 00:10:06.630 --> 00:10:10.330 Irene Palacios: I just did this part of it. So I'm not gonna get a Ted 122 00:10:10.750 --> 00:10:12.919 Irene Palacios: because I didn't do the top one. 123 00:10:13.400 --> 00:10:18.109 Irene Palacios: But hopefully it got water part. And yes, I did have 3 out of 3. 124 00:10:18.900 --> 00:10:22.330 Irene Palacios: Alright. So you first have to decide how you're gonna type in the data 125 00:10:22.350 --> 00:10:24.689 Irene Palacios: type in the table. And remember, you only need 126 00:10:24.830 --> 00:10:29.810 Irene Palacios: the actual data that was collected. You don't need the row or the column tools. 127 00:10:33.960 --> 00:10:39.289 Irene Palacios: And so when you are looking at stack French 128 00:10:39.921 --> 00:10:42.930 Irene Palacios: remember that you can change the 129 00:10:43.880 --> 00:10:47.570 Irene Palacios: names that the call on so you can just go and type 130 00:10:48.420 --> 00:10:51.879 Irene Palacios: click and type directly the names of the columns. 131 00:10:53.640 --> 00:10:57.939 Irene Palacios: and you can size them, make them different sizes. 132 00:10:59.210 --> 00:11:02.349 Irene Palacios: And then what we did is we move on to Stan. 133 00:11:03.070 --> 00:11:07.809 Irene Palacios: and then tables contingency with summary. 134 00:11:08.700 --> 00:11:18.379 Irene Palacios: and we had the response variable and the explanatory variable. The response was the alcohol drinking. But in order to select, you have to 135 00:11:18.480 --> 00:11:21.179 Irene Palacios: press control and then click on the next one. 136 00:11:22.940 --> 00:11:28.189 Irene Palacios: The role label was the work, the explanatory, variable 137 00:11:33.520 --> 00:11:36.269 Irene Palacios: overall labels extend for you. 138 00:11:37.190 --> 00:11:41.440 Irene Palacios: and then we're looking at expect accounts. And then this is the Cha spread 139 00:11:41.490 --> 00:11:43.149 Irene Palacios: test for independence. 140 00:11:44.190 --> 00:11:47.369 Irene Palacios: Press on compute. And that's how we ended up getting that taken 141 00:11:48.340 --> 00:11:50.580 Irene Palacios: alright. I hope this video was helpful 142 00:11:50.640 --> 00:11:53.820 Irene Palacios: to help you out with these contingency tables.